Magnetic resonance imaging, or MRI, is a method by which the location, size, and conformation of organs and other structures of the body may be determined.
In the typical MRI system, a magnetic field is established across a body to align the spin axes of the nuclei of a particular chemical element, usually hydrogen, with the direction of the magnetic field. The aligned, spinning nuclei execute precessional motions around the aligning direction of the magnetic field. For the aligned, spinning nuclei, the frequency at which they precess around the direction of the magnetic field is a function of the particular nucleus which is involved and the magnetic field strength. The selectivity of this precessional frequency with respect to the strength of the applied magnetic field is very short and this precessional frequency is considered a resonant frequency.
In an ordinary MRI system, after the nuclei have been aligned or polarized, a burst of radio frequency energy at the resonant frequency is radiated at the target body to produce a coherent deflection of the spin alignment of the selected nuclei. When the deflecting radio energy is terminated, the deflected or disturbed spin axes are reoriented or realigned, and in this process radiate a characteristic radio frequency signal which can be detected by an external coil and then discriminated in the MRI system to establish image contrast between different types of tissues in the body. MRI systems have a variety of different excitation and discrimination modes available, such as free induction decay ("FID"), spin echo, and continuous wave, as are known in the art.
Two parameters are used to measure the response of the magnetized sample to a disturbance of its magnetic environment. One is T1 or longitudinal relaxation time, the time it takes the sample to become magnetized or polarized after being placed in a external magnetic field; the other is T2, the spin relaxation time, a measure of the time the sample holds a temporary transverse magnetization which is perpendicular to the external magnetic filed. Images based on proton density can be modified by these two additional parameters to enhance differences between tissues.
Hydrogen is usually selected as the basis for MRI scanning because of its prominent magnetic qualities. Hydrogen, having a single proton nucleus, is easily polarized. Further, hydrogen is abundant in water, a major component of the human body. Tissues which have a high content of water, and thus hydrogen and hydrogen protons, are deemed "protonated" and provide strong images during MRI. One disadvantage to hydrogen scanning, however, is that water is a major component of most of the bodily tissues and organs. Therefore, most all of the tissues of the body are imaged, making it difficult to distinguish the various tissues with similar hydrogen content during MRI scanning.
The images formed in magnetic resonance imaging are really a converted visual display of the otherwise invisible radio waves emitted by protons (when scanning for hydrogen atoms) which are detected by the MRI pick-up coil. When scanning for hydrogen atoms, tissue areas which have no hydrogen atoms emit no radio waves, and thus the MR image of this tissue is black. Tissues which have a high hydrogen content, on the other hand, may emit a large amount of radio waves depending on the scanning criteria. Such signals are converted into a correspondingly bright visual display image. Normally, grey scale assignment, based upon the relative energy or signal intensities received from the tissues, is utilized in order that the user may more easily distinguish the various tissues and organs imaged. On these grey scale images, low or no signal is designated as black, and very high signals are assigned a lighter shade of grey or even white.
Occasionally, tissues which are in abundance and create a bright signal may overwhelm the signal emanating from less abundant and differently hydrogenated species or tissue. This may visually mask the latter tissue and obscure a disease process or anatomy. As an example, bone marrow in the adult is very fatty and is very bright on the MR image. Subtle bone marrow pathology such as edema or inflammation may be completely obscured by the signal from the fat. This decreases the sensitivity of P[RI for certain disease processes and creates a problem for the diagnostician.
Various methods have been used in order to try and separate the signals coming from the various tissues of the body and thereby produce more distinct images. One such method involves nullifying the signal received from a certain tissue. This is done by utilizing spin echo and gradient echo presaturation pulse sequences based upon information about subtle differences in the precessional frequency of hydrogen atoms as they associate with fatty versus non-fatty tissues. For example, in order to improve the conspicuity of non-fatty tissues which lie in a background of a fatty tissue, the entire tissue is first subjected to a chemically specific saturation radio pulse. This preparatory pulse essentially effects the hydrogen atoms associated with the fat molecules. These pretreated hydrogen atoms have, in a sense, been briefly deactivated and are not able to emit a useful signal when the actual imaging portion of the pulse sequence commences. The MR image is then created with little or no contribution from the fatty tissue. The resultant image will show the non-fatty tissue against a dark background. This process is called chemically selective presaturation of fat, or fat saturation.
This fat saturation process is unreliable. Because the precessional differences between the fatty and non-fatty tissues are very minute, the technique must be very precise or non-fatty tissues are inadvertently variably saturated themselves. This problem is further compounded by the fact that the local magnetic environment of tissues changes based upon their position relative to the coil; position in the magnetic bore; and position with respect to organs or tissues with different magnetic susceptibilities (e.g. tissue next to bone or tissue next to air). Not only is the immediate magnetic environment important, but the actual geometry of the organ or body part plays a major role in determining the fatty tissue's likelihood of being nullified with the fat saturation technique. For example, fat is more likely to be saturated in the rather cylindrical thigh than in the right angle of the ankle.
Further, interpretive problems can arise in several ways. First, if the fat is not saturated effectively, then pathology can be obscured. Second, if the fat is saturated in only portions of the body part being imaged, then the areas not saturated may be misinterpreted as pathologic tissue. Third, drastic alteration in geometry and magnetic susceptibility which naturally occur in the neck, shoulders and ankle, for example, can lead to inappropriate saturation of non-fatty tissues which are the subject of the examination.
One method occasionally used to improve fat saturation by addressing the above stated limitation of this technique involves placing water bags around the body part being scanned. This technique is useful in that there is improvement in the quality and reliability of the fat saturation technique. This is based on reducing or eliminating the tissue-air interface and by effectively changing the perceived geometry of the body part (e.g. changing the right angle configuration of the ankle to a more favorable cylindrical shape.)
Water, however, is highly protonated and creates a correspondingly bright signal surrounding the fat site. The bright background is a serious disadvantage for this procedure because it is distracting and counteracts the improved visualization produced by using water-filled bags with fat saturation sequences.